RBI's dilemma: sterilisation and capital mobility

How effective have the central banks attempts to sterilise forex inflows been?

With the relentless surge in forex reserves, the conduct of the Reserve Bank of India's (RBI's) monetary policy has been at the forefront of discussion for some time now.

There has been a $ 35 billion accretion in forex reserves during 2003-04 (FY2004) and around $ 3.4 billion in the first week of FY2005 itself. The RBI has been conscious of this spurt and has recently launched a market stabilisation fund (MSF) to neutralise the impact of such forex inflows on the monetary base.

The MSF's primary purpose is to replenish the declining stocks of government securities by creating a separate pool of such inflows.

The securities in the market stabilisation bonds (MSBs) are to be used exclusively for offsetting liquidity injection in the market in lieu of the RBI intervention in the forex market in case of continuing capital inflows (termed sterilisation of capital inflows in economic literature).

This article explores the relative efficacy of sterilisation activities by the RBI as well as its possible impact on domestic interest rates.

But what is sterilisation? When the RBI intervenes in the foreign exchange market, foreign exchange assets (dollar) are exchanged for a liability of the monetary authorities.

For instance, if the monetary authority buys (sells) foreign assets, both sides of the RBI's balance sheet (RBI's assets are forex and domestic assets, while the liabilities are deposits of commercial banks and currency in circulation) will increase (decrease).

This sort of intervention that alters the monetary base is called non-sterilised intervention. The impact on the monetary base can, however, be neutralised by offsetting transactions in domestic assets that have the same magnitude as the transaction in foreign assets.

If the monetary authority makes an open market sale (purchase) of government securities, this will reduce (increase) both the assets and liabilities, offsetting the money-supply effect of the original purchase (sale) of foreign assets.

This type of intervention is termed sterilised intervention. Sterilisation can be achieved through a host of methods such as restrictions that may increase cash reserve ratio (CRR) of commercial banks, or a ceiling on the total credit extended.

The degree of sterilisation is commonly called the sterilisation coefficient (SC). As indicated, the SC is given by the offsetting change in net domestic assets of the central bank by means of open market purchase/sale of government bonds.

It is in the negative and lies between -1 and 0 (if it is -1, it means that the change in net foreign exchange assets arising out of the intervention in the foreign exchange market has been exactly offset by an opposite change in net domestic assets, thus implying full sterilisation).

What is the value of the SC in the Indian context? Table 1 shows the comparative injection and withdrawal of monetary resources into the system by the RBI for the decade ending FY2004. Columns A & B show the liquidity injected through either CRR cuts, or in lieu of the RBI purchase of dollar in the forex market.

In contrast, columns C & D show the liquidity taken out (as indicated by the negative sign) through open market operations (OMOs) and the liquidity adjustment facility mechanism.

The estimated sterilisation coefficient, -0.78, indicates that 78 per cent of the liquidity injected through dollar purchase during FY1994-2004 has been neutralised by the RBI.

However, the coefficient declines to -0.68 if we also take into account the liquidity injection through monetary management by the RBI (that is, CRR cuts).

What is the SC's relevance in the Indian context? Interestingly, it may be noted that the estimated SC for India is lower than the east Asian countries that took recourse to extensive sterilisation in the 1990s to ward off a surge in capital inflows.

For instance, studies indicate that in the case of Korea and Thailand, the sterilisation coefficients (sample period 1988:1997) are in excess of -1 (Korea: -1.01, Thailand: -1.11). An explanation of a coefficient greater than one might be that central banks are sterilising expected capital inflows.

Indonesia had a coefficient of -0.76, which indicates partial sterilisation. Thus, India seems also to have sterilised partially during FY1994-2004.

Irrespective of partial or full sterilisation, the more important thing is the impact on the domestic interest rate. Models in economic literature suggest that the domestic interest rate is a function of domestic and international monetary conditions.

International monetary conditions are in turn determined by capital mobility "" the higher the mobility, the more is the importance of external factors. In a similar vein, as capital mobility declines, the domestic interest rate is determined more by domestic monetary conditions.

In the context of the east Asian economies, given that the capital accounts in these countries have been literally open, the sterilisation policies (domestic monetary conditions were not relevant) have been less ineffective as far as the impact on domestic interest rates is concerned.

Coming back to the Indian context, as the estimates suggest, only 78 per cent of the forex assets have been offset by a change in domestic credit over the sample period. For an inflow of capital, this means that there has been an increase in the monetary base. All other things remaining unchanged, this may have, therefore, resulted in lower domestic interest rates (domestic factors outweighed international factors).

In particular, the M3 or money supply growth in FY2004 has been 15.8 per cent, as against the targeted 14 per cent by the RBI. This has been the result of an Rs 1413 billion increase in forex assets of the RBI, which was offset only by a decline of Rs 962 billion domestic assets of the apex bank during FY2004.

Sterilisation in the Indian context has, therefore, been successful, so far as its impact on domestic interest rates is concerned. However, as mentioned above, such optimism should be viewed against the fact that the capital account in India is still not fully liberalised.

Otherwise, there may have been a virtual loss of monetary autonomy as was the case with the east Asian countries (with a coefficient greater than -1, monetary policy was subject to an exchange rate rule rather than a money growth rule).

Finally, a note on the composition of capital inflows into the East Asian countries vis-à-vis India. Table 2 reports capital inflow by foreign direct investment (FDI), portfolio flows and other flows.

With the exception of Malaysia "" which had strong FDI flows pre-crisis "" the largest category of inflow was largely through bank lending. It is this bank-lending channel that may be inflationary if not sterilised as the inflow has a direct impact on the money multiplier process.

As against this, the composition of capital inflows for India also reveals the domination of bank capital flows (primarily, NRI deposits), though not to the same extent as east Asia.

No wonder the RBI has been successively reducing incentives for NRI deposits in terms of interest arbitrage opportunities (on April 17, 2004, the RBI lowered the interest rate ceilings on non-resident Indian deposits for the fourth time in less than 12 months).

So what is the RBI's dilemma? With the progressive liberalisation of the capital account, there will possibly be increased sterilisation activities by the bank.

However, if the experiences of the east Asian countries are to be believed, the apex bank will be hard pressed to ensure the success of such sterilisation. But whether to liberalise the capital account or not will remain a million-dollar question.

(The author is with ICRA Limited. The views expressed are personal)